Dashboards

Dashboards turn the telemetry every Hanzo service emits into product analytics and observability views: usage over time, cost by product and model, latency, error rates, and live health. They are the read surface over your platform's metrics.

Analytics and Observability

Two families of dashboards share one home in the Console:

• Product analytics -- volume, active users, adoption, and cost trends per product, model, and organization.
• Observability -- request traces, latency percentiles, error rates, and service health across your clusters.

Every panel is scoped to your organization and reads real metrics. Usage is attributed the same way Billing meters it, so what you see and what you are charged reconcile.

Reading Metrics via the API

Aggregated metrics are available under /v1 on the gateway, so you can pull the same numbers the dashboards render into your own tools. Authenticate with an API key (hk-...):

# Daily aggregated usage and cost
curl "https://api.hanzo.ai/v1/analytics/usage?period=daily" \
  -H "Authorization: Bearer hk-..."

For the exact metric endpoints and query parameters, see the API reference.

Related

• Console -- where dashboards live
• Billing -- cost and usage, metered from the same telemetry
• API -- pull metrics programmatically
• Gateway -- the entry point that emits request telemetry